Photogrammetric Surveys ACQUISITION AND DATA PROCESSING

3.3 Photogrammetric Surveys

The equipment used for image capture was composed of: - A digital camera Canon EOS 5D type SLR - A digital camera Sony Bridge H50 - A tripod. Images were taken to model briefly the outside of the “Seigniory”. The PhotoModeler software was used to model and texture the 3D model. This survey method was also tested on the frame. Lighting facilities were established to easily materialize this stage. Ramps of neon tubes were arranged on both sides of each photographed area. With this method, geo-referencing is harder than with a 3D scanner and above all, it is difficult to control field data acquisition; this can be considered as the disadvantages of such a method. 3.4 Total Station Survey Another method coming from conventional survey was also tested in this study. The beams were identified through a total station Leica “TS02” Total Station, taking measurements on the ends and edges of them. Compared to the point clouds acquired by 3D scanner, the number of data is very low but the points are all significant. An exhaustive survey is much longer as with a 3D scanner and the necessarily limited number of measured points does not allow a detailed study of the structure. While it is still easy to find the beams having no distortion, those which are deformed over time are more problematic and need a multiplication of measures. Lighting is also very delicate, and therefore it remains difficult for the operator to identify correctly the end of the beams, which can be a source of error. The advantage of this method is that it is possible to codify the measures and thus automate, subsequent processing to CAD drawing. This feature is not included in the 3D scanner measures. While time is long on field measurement, processing time is much shorter. This method is recommended for a brief survey of the structure, which is not the case here. Since the structure will be changed soon, the method providing a maximum data of the current state was preferred. This data can constitute interesting archives. They can be processed afterwards if necessary, and retrieving information can be essential when they are not even more visible, nor accessible. The choice of using a TLS finds here a good justification in the context of this study. 3.5 Panoramic Image Capture The material used for the panoramic shots was: - A camera Canon EOS 5D , type SLR , - A spherical panoramic head Manfrotto 303+ , - A tripod. During the study, a new opportunity that was not originally planned to see the structure was tested: the virtual tour from panoramic photographs. The formation of an assembled panoramic photograph requires a large number of photos using a spherical panorama head. In general we must ensure an overlap between photographs of about 30 minimum and we cannot use photographs with too pronounced contrasts differences. But with the environmental conditions in the attic lighting difficulties we used recoveries of approximately 50 . So it was easier to rectify underexposed or overexposed photographs and also to focus on different focal planes in the final panoramic photo. The assembly and photo rectification was performed on the Autopano Giga software. The positions of different points of view were chosen according to several criteria: - to ensure the inter-visibility between adjacent positions, - to link lower and upper level at each level in the panoramic tour, - outside, it is necessary to view the “Seigniory” from different angles and in its environment. By following these criteria, the virtual tour created allows the user to move from one point of view to another one, as he would do naturally in the real world. Each picture was taken with a gap of 24° horizontally with approximately 50 overlap and with a lens focal length of 24 mm. That means 15 photos to cover 360°. A complete panoramic photograph requires in this case 5 series of 15 photographs taken with vertical angles of -30°, 0°, +30° and +60°, then one last image at +90°. This brings us to a total of 61 photographs.

4. MODELLING